Computer Performance Appraisal System Based on Balanced Scorecard under the Internet Background

Abstract

In the new era of rapid development of Internet information technology and data processing technology, although the emergence of Internet technology has brought changes that cannot be underestimated in various fields of society, the computer performance evaluation system based on the Internet in the era of big data will definitely follow Technology trends in the new era. The balanced scorecard is one of the most commonly used financial performance appraisal tools in academia. The research purpose of this article is to study the innovation of enterprise computer performance appraisal management system in the current Internet era with the continuous development of big data technology. This paper first uses a neural network algorithm to build a model for it. Through this algorithm combined with a questionnaire survey method to screen the evaluation indicators of the computer performance evaluation system of Internet companies in the eastern coastal area, and finally build a BP model through the enterprise computer performance evaluation system. Model calculation samples, after calculation, the computer performance analysis of the target company can be obtained, and on the basis of summarizing the computer performance evaluation system and the balanced score card of the enterprise in the Internet era, the in-depth understanding and conclusion can be drawn. After studying the setting of performance appraisal system indicators, the setting of personal growth indicators is insufficient, the lack of effective cost control and the setting of computer performance indicators have not yet formed a complete system problem. This paper combines the actual situation of computer performance appraisal of Internet companies in the eastern coastal areas, applies the balanced scorecard to set up a management index system for the company’s computer performance appraisal, and uses the analytic hierarchy process to analyze the weights of the various indicators of the corporate computer performance appraisal balanced scorecard. Compare the evaluation results with the designated enterprise computer performance management evaluation standards. Experimental research results show that the balanced scorecard can effectively improve the efficiency of the computer performance appraisal system, and the computer performance appraisal can be clearer, and based on the Internet background and big data technology, the performance appraisal system can be made clearer.

Full text

Journal of Physics: Conference Series
PAPER • OPEN ACCESS
Computer Performance Appraisal System Based
on Balanced Scorecard under the Internet
Background
To cite this article: Haiyan Wu and Yongjun Qi 2021 J. Phys.: Conf. Ser. 1881 032012
View the article online for updates and enhancements.
You may also like
Designing Performance Measurement For
Supply Chain's Actors And Regulator
Using Scale Balanced Scorecard And
Data Envelopment Analysis
Elisa Kusrini, Subagyo and Nur Aini
Masruroh
-
Development of a performance
measurement tool for an agricultural
enterprise using BSC and QFD models
R Dwivedi, S Chakraborty, A K Sinha et al.
-
Prerequisites for the implementation of a
balanced scorecard in strategic forest
management
Y Stepanova, I Sibiryatkina, Y Busarina et
al.
-
This content was downloaded from IP address 45.114.243.109 on 19/01/2023 at 18:01

Content from this work may be used under the terms of the Creative Commons Attribution 3.0 licence. Any further distribution
of this work must maintain attribution to the author(s) and the title of the work, journal citation and DOI.
Published under licence by IOP Publishing Ltd
The 2nd International Conference on Computing and Data Science (CONF-CDS 2021)
Journal of Physics: Conference Series 1881 (2021) 032012
IOP Publishing
doi:10.1088/1742-6596/1881/3/032012
1
Computer Performance Appraisal System Based on Balanced
Scorecard under the Internet Background
Haiyan Wu
1
and Yongjun Qi
2,*
1
Management School, South China Business College Guangdong University of
Foreign Studies, Guangzhou, China
2
Faculty of Megadate and Computing, Guangdong Baiyun University, Guangzhou,
China
*Corresponding author e-mail: qiyongjun@baiyunu.edu.cn
Abstract. In the new era of rapid development of Internet information technology and
data processing technology, although the emergence of Internet technology has
brought changes that cannot be underestimated in various fields of society, the
computer performance evaluation system based on the Internet in the era of big data
will definitely follow Technology trends in the new era. The balanced scorecard is one
of the most commonly used financial performance appraisal tools in academia. The
research purpose of this article is to study the innovation of enterprise computer
performance appraisal management system in the current Internet era with the
continuous development of big data technology. This paper first uses a neural network
algorithm to build a model for it. Through this algorithm combined with a
questionnaire survey method to screen the evaluation indicators of the computer
performance evaluation system of Internet companies in the eastern coastal area, and
finally build a BP model through the enterprise computer performance evaluation
system. Model calculation samples, after calculation, the computer performance
analysis of the target company can be obtained, and on the basis of summarizing the
computer performance evaluation system and the balanced score card of the enterprise
in the Internet era, the in-depth understanding and conclusion can be drawn. After
studying the setting of performance appraisal system indicators, the setting of personal
growth indicators is insufficient, the lack of effective cost control and the setting of
computer performance indicators have not yet formed a complete system problem.
This paper combines the actual situation of computer performance appraisal of
Internet companies in the eastern coastal areas, applies the balanced scorecard to set
up a management index system for the company's computer performance appraisal,
and uses the analytic hierarchy process to analyze the weights of the various indicators
of the corporate computer performance appraisal balanced scorecard. Compare the
evaluation results with the designated enterprise computer performance management
evaluation standards. Experimental research results show that the balanced scorecard
can effectively improve the efficiency of the computer performance appraisal system,
and the computer performance appraisal can be clearer, and based on the Internet
background and big data technology, the performance appraisal system can be made
clearer.

The 2nd International Conference on Computing and Data Science (CONF-CDS 2021)
Journal of Physics: Conference Series 1881 (2021) 032012
IOP Publishing
doi:10.1088/1742-6596/1881/3/032012
2
Keywords: Internet, Big Data Technology, Balanced Scorecard, Computer,
Performance Appraisal System
1. Introduction
The balanced scorecard is an epoch-making strategic management performance evaluation tool
proposed by the famous American management masters David P. Norton and Robert S. Kaplan on the
basis of the successful experience of large-scale enterprise performance. It is called an effective
performance appraisal management tool [1]. It is not only widely used by companies, but also highly
recognized by non-profit organizations and government and military institutions [2]. With the
development of economic globalization, the competition of enterprises is more reflected in the
competition of talents. How to conduct effective human resource management for them, implement
scientific and reasonable performance appraisal, and maximize their potential is of far-reaching
significance for the development of enterprises and employees [3].
From the perspective of the enterprise, the scientific and reasonable performance appraisal can
form comprehensive and accurate appraisal results. This provides data accumulation and fundamental
basis for the company to determine comprehensive assessments such as personnel adjustments, salary
changes, and advanced training. At the same time, it also laid the foundation for the company to
continuously introduce talents and reduce the turnover rate [4, 5]. From the perspective of employees,
objective and fair performance appraisal can not only effectively stimulate their work enthusiasm and
initiative, but also meet their material and spiritual needs to a certain extent, and maximize their
realization. Self-worth. In addition, the evaluation of the work performance of the employees in the
appraisal, affirmed that the employees have paid, but also pointed out the shortcomings, which can
guide the employees to continuously improve performance, further pursue work results, and obtain
personal career development and promotion [6, 7].
In the operation of some modern enterprises, due to the lack of scientific evaluation standards and
reasonable management systems within the enterprise, coupled with the influence of other special
factors, a considerable number of employees will think that performance evaluation management is
mainly for monitoring and management of themselves. Lead to the enthusiasm and enthusiasm of the
employees themselves [8]. It is extremely easy to affect the sense of belonging of the company, and on
the other hand, it causes employees to lack sufficient sense of security in their daily work, which will
not only affect the performance appraisal work, but also affect the healthy development of related
companies [9,10].
2. Method
2.1. Principle Model and Algorithm
Let (Xk, Dk), k=1,2,...,n be the input and output sample data, for which Xk=(x1,x2,...,xm)T,
Dk=(d1,d2,...,dp)T . Taking Xk as the input of the network, under the action of the connection right,
the actual output of the network Yk=(y1,y2,...yp)T can be obtained. Assuming that the connection
weight of neuron i to j is wij, the adjustment amount of the weight is:
 
 


  (1)
The above formula is the learning rate, is the partial derivative of the error function to the input of
neuron j, Vi is the output of the i-th neuron, and the error correction learning algorithm is a very
important learning rule in the neural network.

The 2nd International Conference on Computing and Data Science (CONF-CDS 2021)
Journal of Physics: Conference Series 1881 (2021) 032012
IOP Publishing
doi:10.1088/1742-6596/1881/3/032012
3
In recent years, neural networks have been widely used in many fields, such as computer vision,
intelligent robot fault detection, real-time translation, business management, market analysis and other
fields.
2.2. Algorithms and Steps of the BP Model for Corporate Performance Appraisal
First, select the k-th sample input sample and the corresponding expected input.
Using the input layer to modify the connection authorization value is to use the xi(k) of each
neuron in the hidden layer and each neuron in the input layer to modify the connection value formula
as follows:

 
 (2)
The formula for calculating the global error E is as follows:
 
  



 (3)
When E< or the number of learning times is greater than the set maximum number of times M, the
calculation of the algorithm is ended, otherwise the next learning sample and the corresponding
expected output are selected.
2.3. Analysis of the Calculation System of the Classification Weight of Employee Performance
Appraisal
When the number of classifiers in the supervision integration system does not reach the maximum
value, use the latest data block to build a new classifier and join the integration system. The error of
the new classifier is calculated by cross-validation, and the weight and mean square of the classifier
The error is inversely proportional, and the mean square error of the random classifier is:
 !"#$"#
% (4)
& $
 !
 &' 
()($*%&+
,)-./%-01  (5)
Therefore, the mean square error can be calculated for the classifier in the integrated system on the
latest data block, and the weight of the i-th base classifier on the J-th sample of the data blocks is:
2&' 
3.453.4678 (6)
The classification error rate of the integrated classifier is the proportion of the number of correctly
classified instances to the total number of instances. The calculation formula is as follows:
99 
:&;)&<=>? '@
:&;)&<=
& (7)
2.4. Test and Application of the Effect of Computer Performance Assessment Model
Obtain a well-trained BP neural network. According to the calculated BP network connection weight
and threshold, the performance evaluation effect is tested and applied in practice. Standardize and
preprocess the measurement values of the enterprise's comprehensive computer performance appraisal
as the input of the trained BP model. After calculation, the performance of the target company’s
computer performance appraisal can be obtained, and the evaluation result can be compared with the
specified corporate computer performance appraisal performance appraisal standard to determine the
target company’s performance category and level.

The 2nd International Conference on Computing and Data Science (CONF-CDS 2021)
Journal of Physics: Conference Series 1881 (2021) 032012
IOP Publishing
doi:10.1088/1742-6596/1881/3/032012
4
3. Experiment
3.1. Determine the Experimental Object
Based on the design of this research, the research variables are the analysis and research of enterprise
interconnection big data capabilities and enterprise computer performance appraisal system. In order
to obtain high-quality samples to ensure the real validity of residence, this research selects M
enterprises as the survey samples. Statistics based on the Internet and big data technology show that
the enterprise database in the eastern coastal area has already received about 300 enterprises, and the
number of enterprises will further increase in 2020, due to the application and industrial development
of the eastern coastal area in the country leading Therefore, the sample data has a certain degree of
representativeness. This article uses questionnaire surveys to survey the eastern coastal areas. The first
is to conduct a questionnaire survey for middle and high-level companies participating in big data
courses. The second method is to send emails to some companies with big data time experience or IT
management. Experienced middle and senior managers conduct a questionnaire survey.
3.2. Employee Performance Appraisal Algorithm Model Construction
K-means clustering implements employee evaluation as follows:
First, for the construction of the performance appraisal system, k is set to 4 categories for analysis,
and n is composed of 12 data sources, and finally K clusters are output to meet the minimum change
value of the performance appraisal calculation formula.
By summarizing the datum points of the data of the data set for classification, calculating the
clustering interval D obtained from the performance data training set, then the D points are about the
class, where:
  ,$/A/B/C/D/E/F1G  ,$/A/B/C1G (8)
According to the following Euclidean distance method, the cluster centers are involved as shown in
(9):
H?I'/J'@  KI' J'
;7
' (9)
Obtain the calculated values of clusters in the enterprise computer performance appraisal system
cluster by refreshing, and calculate the data attributes and relative average values in all clusters;
Calculate the criterion function according to the general formula of K-means for clustering. (10) The
formula is as follows:
L    ((IJ'((
MNO7

' (10)
Finally, until performance does not change
Finally, through the analysis of the employee evaluation link of the quantitative performance
appraisal system implemented this time, the realization of the balanced scorecard and clustering
algorithm is not limited to a certain algorithm, assuming that we can achieve many kinds of Clustering
method, so we have some limitations in the realization of multiple clustering methods for an algorithm
to explain the division of Sydney Harbour, so this algorithm must integrate multiple clustering
techniques to play its role.
According to the 100 sample data collected from the prediction and evaluation data, 5 groups of
representative data are selected for rationality verification. The 5 groups of enterprise employee
performance appraisal system evaluation data are shown in Table 1:

The 2nd International Conference on Computing and Data Science (CONF-CDS 2021)
Journal of Physics: Conference Series 1881 (2021) 032012
IOP Publishing
doi:10.1088/1742-6596/1881/3/032012
5
Table 1. Data Analysis of Computer Performance Appraisal System for Group of Employees
Name Employee Performance Appraisal And Evaluation System
An
employee
Work
performance
Work
efficiency
Communication
skills Teamwork
Total
B employee
7 8 8 9 32
C employee
8 8 6 9 31
D employee
9 9 7 8 33
E employee 7 8 9 9 33
4. Results
It can be seen from Figure 1 that the investigators’ company working hours (less than 3 years, 3-5
years, 5-10 years, and more than 10 years) and other different working hours were investigated. The
work ability of employees has a greater impact on the management of corporate performance.
Relevant survey data show that the work of employees has seriously affected their work performance.
Due to high work pressure, long working hours, long working hours, and long-term concentration and
tension, psychological pressure problems related to employees’ emotions, pressure, communication,
interpersonal relationships, depression, fatigue, etc. are increasing, which not only cause physical and
mental stress to employees Greatly hurt, and seriously affect the work performance of employees,
reducing the efficiency of the company's performance appraisal system for employees.
Figure 1. Sample survey of M company employees' working hours analysis diagram
In the context of the Internet era, an enterprise's human resource performance appraisal system is
one of the important content of an enterprise to improve its breeding ability. It is the collection of data.
The data collection channel is not a single one, but is reflected in multiple aspects. In the resource
management department, the human resources management department will make the final statistical
classification. The data is organized to facilitate managers to have a clear understanding of personnel
changes, and lay a solid foundation for future work. The data of human resources dynamic changes
follow the personnel Changes and constant changes, managers should analyze and judge the data
before using it, in order to maximize its value in the human resource performance management
system.
32%
41% 37%
49%
0%
10%
20%
30%
40%
50%
60%
Less
than 3
years
3-5
years
5-10
years
More
than
10
years
percentage

The 2nd International Conference on Computing and Data Science (CONF-CDS 2021)
Journal of Physics: Conference Series 1881 (2021) 032012
IOP Publishing
doi:10.1088/1742-6596/1881/3/032012
6
Table 2. Descriptive Statistics of Each Dimension of the Balanced Scorecard in Computer
Performance Appraisal Management
Numbering
Process
Dimension Item
Sort Minimum
Maximum
Mean Standard
Deviation
09 Learning growth
dimension 1 2.00 6.00 3.5735 0.7315
13 Financial
dimension 2 3.00 4.00 6.2317 0.6310
15 Internal process
dimension 3 5.00 7.00 4.2519 0.7257
17 Customer service
dimension 4 1.00 3.00 1.5317 0.2319
As shown in Table 2 for the data displayed by the descriptive statistics of each dimension of the
balanced scorecard in the computer performance appraisal management, the descriptive statistical
analysis of the internal process dimension indicators of the sample is carried out, and the indicators are
considered to be of certain importance. Among them, the average value of the "financial dimension"
index is the highest, which can be understood as whether the financial management structure can be
appropriately adjusted according to the changes in the current talent market demand, whether financial
management can quickly adapt to the needs of local economic and social development, and whether it
is scientific and reasonable. The key to improving the level of corporate performance, promoting the
development of corporate connotation, and achieving the strategic goals of the college lies. In addition,
the "learning growth dimension", "internal process dimension" and "customer service dimension"
indicators have higher average scores.
Synthesize the above data, use the BP model to analyze the corporate performance appraisal system,
use the K-means algorithm to analyze the calculations based on the balanced scorecard, and
comprehensively study the processing distribution diagram of the corporate employee performance
appraisal analysis as shown in Figure 2.
Figure 2. Distribution diagram of analysis and processing of employee performance appraisal based
on BP model of balanced scorecard
5. Conclusion
The balanced scorecard is not only a performance appraisal management method, but also embodies a
balanced performance management idea. By balancing financial and non-financial indicators, the
balanced scorecard expands the focus of corporate financial managers on short-term and long-term
financial management performance appraisal, and cannot ignore key points. In the era of Internet big
data, computers use digital information technology in the system of corporate performance appraisal
based on the balanced scorecard. It is necessary to informatize various data within the enterprise and
use the data to lead the performance appraisal work. The arrival of the Internet big data era means
98.23%
0.70% 1.07%
Can handle
Unprocessable
other

The 2nd International Conference on Computing and Data Science (CONF-CDS 2021)
Journal of Physics: Conference Series 1881 (2021) 032012
IOP Publishing
doi:10.1088/1742-6596/1881/3/032012
7
comprehensive innovation in enterprise performance resource management. In the context of today's
era, how to effectively grasp the opportunities of big data application is an important way to realize the
innovation of enterprise performance appraisal management, and it is also an important issue that
needs to be explored in the field of enterprise performance resources.
References
[1] Muda I Msi E E S Yahya I et al. Performance Audit and Balanced Scorecard Perspective.
International Journal of Civil Engineering and Technology, 2018, 9(5):1321-1333.
[2] Terziev V Stoyanov E Georgiev M. МОДЕЛЪТ БАЛАНСИРАНА КАРТА ЗА ОЦЕНКА
КАТО ИНСТРУМЕНТ ЗА УСЪВЪРШЕНСТВАНЕ НА УПРАВЛЕНИЕТО В
РАБОТАТА НА НАЦИОНАЛНАТА СИСТЕМА ЗА ОТБРАНА И СИГУРНОСТ (The
Balanced Scorecard Model As a Tool for Improvement of the National Defense System
Management). Social Science Electronic Publishing, 2018, 38(1):1028–1035.
[3] Wiraeus D Creelman J .Agile Strategy Management in the Digital Age (How Dynamic
Balanced Scorecards Transform Decision Making, Speed and Effectiveness) ||From
Industrial- to Digital-Age-Based Strategies. 2019, 10.1007/978-3-319-76309-5 (Chapter
2):23-44.
[4] Cakasana R Sunardi S. PENGGUKURAN KINERJA RANTAI PASOK BERBASIS LEAN
DAN GREEN DENGAN PERSPEKTIF BALANCES SCORECARD DI PT. XYZ.
JUMINTEN, 2020, 1(1):87-95.
[5] Rachman E P Sumardi S Munizu M. Performance Evaluation Analysis of Tirta Jeneberang
PDAM in Gowa Regency Based on Balance Scorecard. HASANUDDIN JOURNAL OF
BUSINESS STRATEGY, 2019, 1(2):25-35.
[6] Limasan J Ahmad R Bandar N F A. Application of performance appraisal system (PAS)
attributes among first rater officers (FRO's): Issues and Implication. Journal of Cognitive
Sciences and Human Development, 2019, 5(2):66-77.
[7] He W Chen L Liu W. Does new performance appraisal system (EVA) affect earnings
management? Nankai Business Review International, 2020, 11(2):191-216.
[8] Taddese G E. ACADEMIC STAFFS PERCEPTIONS TOWARD PERFORMANCE
APPRAISAL SYSTEM IN HIGHER EDUCATION (CASE OF SAMARA UNIVERSITY).
International Journal of Advanced Research, 2020, 8(3):967-976.
[9] Khan M A Hussain A Khan M H. The Importance of Organizational Justice, Appraisal Purposes
and Employee Satisfaction in Performance Appraisal System in Academic Sector of Pakistan.
Journal of Accounting and Finance in Emerging Economies, 2020, 6(1):191-200.
[10] Lim M Hastie T. glinternet: Learning Interactions via Hierarchical Group-Lasso Regularization.
Journal of Computational & Graphical Statistics, 2018, 24(3):627-654.